deep sequencing reveals a novel mir-22 regulatory network with therapeutic potential ... ·...

Therapeutics, Targets, and Chemical Biology

Deep Sequencing Reveals a Novel miR-22Regulatory Network with Therapeutic Potentialin RhabdomyosarcomaFrancesca Bersani1,2, Marcello Francesco Lingua1,2, Deborah Morena1,2,Valentina Foglizzo1,2, Silvia Miretti3, Letizia Lanzetti1,4, Giovanna Carr�a5,Alessandro Morotti5, Ugo Ala6, Paolo Provero6, Roberto Chiarle2,6,7, Samuel Singer8,Marc Ladanyi9, Thomas Tuschl10, Carola Ponzetto1,2, and Riccardo Taulli1,2

Abstract

Current therapeutic options for the pediatric cancer rhabdo-myosarcoma have not improved significantly, especially formetastatic rhabdomyosarcoma. In the current work, we per-formed a deep miRNA profiling of the three major humanrhabdomyosarcoma subtypes, along with cell lines and normalmuscle, to identify novel molecular circuits with therapeuticpotential. The signature we determined could discriminaterhabdomyosarcoma from muscle, revealing a subset of mus-cle-enriched miRNA (myomiR), including miR-22, which wasstrongly underexpressed in tumors. miR-22 was physiologicallyinduced during normal myogenic differentiation and was tran-scriptionally regulated by MyoD, confirming its identity as amyomiR. Once introduced into rhabdomyosarcoma cells, miR-

22 decreased cell proliferation, anchorage-independent growth,invasiveness, and promoted apoptosis. Moreover, restoringmiR-22 expression blocked tumor growth and prevented tumordissemination in vivo. Gene expression profiling analysis ofmiR-22–expressing cells suggested TACC1 and RAB5B as pos-sible direct miR-22 targets. Accordingly, loss- and gain-of-func-tion experiments defined the biological relevance of these genesin rhabdomyosarcoma pathogenesis. Finally, we demonstratedthe ability of miR-22 to intercept and overcome the intrinsicresistance to MEK inhibition based on ERBB3 upregulation.Overall, our results identified a novel miR-22 regulatory net-work with critical therapeutic implications in rhabdomyosar-coma. Cancer Res; 76(20); 6095–106. �2016 AACR.

IntroductionmiRNAs are small noncodingRNAmolecules that regulate gene

expression at the posttranscriptional level (1, 2). In the pastdecade, the critical role of miRNAs has been reported in a wide

range of biological functions including cell proliferation, migra-tion, apoptosis, lineage commitment, and differentiation (3, 4).Thus, it is not surprising that miRNAs are frequently dysregulatedin cancer, where they can act either as oncogenes or oncosup-pressors depending on their downstream target genes (5). Fur-thermore, miRNA signatures have been successfully used toclassify human cancers and to discover novel attractive targetsfor therapeutic intervention (6).

Rhabdomyosarcoma is the most common soft tissue sarco-ma of childhood and the third most common in young adults(7). Histopathologic classification includes three major sub-types: embryonal (ERMS), alveolar (ARMS), and pleomorphic(PRMS). ERMS is more frequent, genetically heterogeneous,and associated with a better prognosis (8–10). Conversely,ARMS is less common and more aggressive, with a worseoutcome. A dismal clinical prognosis is genetically ascribed tothe PAX3/7-FKHR translocation present in more than 80% ofthe cases (11). Notably, rhabdomyosarcoma cells are positivefor myogenic markers and resemble normal muscle progenitorsbut are unable to complete the differentiation program (12).We have previously showed that the muscle-enriched miRNAs,miR-1 and miR-206 (myomiRs), reprogram the rhabdomyo-sarcoma expression profile toward that of a normal muscle, in aprocess involving the post-transcriptional regulation of hun-dreds of genes, among which are MET, G6PD, ACTL6A, andSMYD1 (13–15).

In this work, we used a next-generation miRNA sequencingapproach (NGS) on a large panel of human rhabdomyosarcoma

1Department of Oncology, University of Turin, Orbassano, Turin, Italy.2CeRMS,Center for Experimental ResearchandMedical Studies,Turin,Italy. 3Department of Veterinary Science, University of Turin, Gru-gliasco, Italy. 4Candiolo Cancer Institute, Candiolo, Turin, Italy.5Department of Clinical and Biological Sciences, University of Turin,Orbassano, Italy. 6DepartmentofMolecular BiotechnologyandHealthSciences, University of Turin, Turin, Italy. 7Department of Pathology,Boston Children's Hospital and Harvard Medical School, Boston, Mas-sachusetts. 8Departmentof Surgery,Memorial SloanKetteringCancerCenter, New York, New York. 9Department of Pathology, MemorialSloan Kettering Cancer Center, New York, New York. 10Department ofRNA Molecular Biology, Howard Hughes Medical Institute, The Rock-efeller University, New York, New York.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

F. Bersani and M.F. Lingua contributed equally to this article.

Current address for V. Foglizzo: The Francis Crick Institute, Mill Hill Laboratory,The Ridgeway, London, United Kingdom.

Corresponding Authors: R. Taulli, Department of Oncology, University of Turin,Via Santena 5, Turin 10126, Italy. Phone: 39 011 633 4566; Fax: 39 011 633 6887;E-mail: [email protected]; and C. Ponzetto, [email protected]

doi: 10.1158/0008-5472.CAN-16-0709

�2016 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 6095

on June 8, 2020. © 2016 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 28, 2016; DOI: 10.1158/0008-5472.CAN-16-0709

http://cancerres.aacrjournals.org/

primary tumors, including the three major subtypes, cell lines,and normal muscle tissues, to identify novel miRNA-regulatorycircuits involved in rhabdomyosarcoma pathogenesis. ThemiRNA signature clearly distinguished malignant tissues fromnormal skeletalmuscle and revealed a strong reduction ofmiR-22andmiR-378 in rhabdomyosarcoma. However, only the rescue ofmiR-22 exerted a very potent oncosuppressor function, interferingwith the transformed properties of rhabdomyosarcoma cells bothin vitro and in vivo. Gene expression profiling of miR-22–expres-sing rhabdomyosarcoma cells suggested TACC1 and RAB5B astwo critical miR-22 targets, while ERBB3 emerged only upon thetreatment of mutant NRAS–positive cells with MEK inhibitors.Altogether, our NGSmiRNA sequencing effort uncovered a novelmiR-22 oncosuppressor regulatory circuit that opposes rhabdo-myosarcoma tumor growth and interferes with the resistance toMEK inhibition.

Materials and MethodsCell lines

Embryonal (RD18, CCA, HTB82, TE671, indicated as Myosar-coma_TE) and alveolar (RH4, RH30) rhabdomyosarcoma celllines were provided by Dr. Pier-Luigi Lollini (University of Bolo-gna, Bologna, Italy). The pleomorphic cell line RMS-559 wasobtained from Samuel Singer's laboratory. HTB82 and TE671 celllines were originally obtained from ATCC; RH30 and RH4(RH41) were originally obtained from DSMZ; CCA and RD18cell lines were originally stabilized in Pier-Luigi Lollini's labora-tory. C2C12 myoblasts were originally obtained from DSMZ.Satellite cells, RD18 NpBI-206 cells, RD18 NpBI-206AS cells, andNIH 10T1/2 NpBI-MyoD cells were described previously (13–15).Rhabdomyosarcoma cell lines,NIH10T1/2 cells, satellite cells, andmyoblasts were grown as described previously (13). RD18,HTB82, TE671, RH4, and RH30 cell lines were routinely authen-ticated (every 6 months) by short tandem repeat (STR) analysis.CCA cell line, for which STR profile is unknown, was authenti-cated by sequencing of the KRAS Q61L mutation.

PatientsPrimary human tumors of embryonal, alveolar, and pleomor-

phic histology (or their RNA) and muscle tissues were obtainedfrom Memorial Sloan Kettering Cancer Center (New York, NY),with informed consent prior to the inclusion in the study andwithobscured identity, according to the recommendations of theInstitutional Review Board of the Memorial Sloan KetteringCancer Center. For all ARMS samples, the presence of thespecific fusion transcripts was confirmed by RT-PCR. Of the14 rhabdomyosarcomas included in this study, 10 had previ-ously been extensively analyzed by gene expression profiling,confirming subtype-specific signatures (16). Normal cell con-tamination of the processed specimens was reviewed andassessed to be less than 20%.

Small RNA isolation and library generationRNA from cultured cells, freshly frozen, and optimal cutting

temperature compound–embedded tissues was extracted usingTRIzol (Invitrogen). RNA from formalin-fixed, paraffin-embed-ded tissues was isolated with MasterPure RNA Purification Kit(Epicentre Biotechnologies). Despite a different yield of totalRNA, the miRNA expression profiles of all types of samples arewell correlated across various histologic subtypes. cDNA library

preparation was performed as described previously (17). A briefexplanation can be found in Supplementary Materials and Meth-ods. Sequencing was performed at Memorial Sloan KetteringCancer Center (New York, NY) and raw data are deposited onSRA platform, ID PRJNA326118. Computational analysis of theraw data was done in collaboration with Mihaela Zavolan'slaboratory (University of Basel, Basel, Switzerland).

Lentiviral vectors and siRNAsNpBI-22 and NpBI-378 vectors were generated as described

previously (13). Vectors and si/shRNAs are detailed in Supple-mentary Materials and Methods.

Northern blot analysisNorthern blot analysis was performed as described previously

(13). 32P-labeled DNA oligos are listed in Supplementary Materi-als and Methods.

Microarrays and data analysisAffymetrix Human GeneChip Gene ST 1.0 arrays (Affymetrix)

were hybridized at the Cogentech core facility according to thestandard Affymetrix protocols. The array data were analyzed withthe Partek Genomics Suite. All genes showing differential expres-sion between the two experimental conditions found to besignificant by ANOVA were then subjected to unsupervised hier-archical clustering. Microarray data were deposited under seriesGSE83805.

Sensor vector generation and reporter assaysGFP/luciferase reporter vectors were cotransfected with syn-

thetic pre-miRNA (miRNA precursor hsa-miR-22, PM10203,Thermo Fisher) in HEK 293T cells. For validation of MyoD-binding sites, luciferase reporter vectors were transfected in NIH10T1/2 NpBI-MyoD cells. Detailed information is summarized inSupplementary Materials and Methods.

Real-time PCR analysisTaqMan miRNA Assays (Applied Biosystems) were used for

relative quantification of mature miRNAs. miR-16 was used tonormalize the results. Gene expressionwas evaluated as describedpreviously (15), using primers listed in Supplementary Materialsand Methods. HuPO was used to normalize the results.

Cell proliferation assay, cell-cycle analysis, and assessment ofapoptosis

For proliferation assay, cells were plated in 24-well plates at adensity of 5 � 103 cells per well. Proliferation was evaluated byCellTiter-Glo (Promega) following the manufacturer's instruc-tions. Cell-cycle analysis and apoptosis were performed asdescribed previously (13).

Anchorage-independent cell-growth, invasiveness, and scratchwound assays

Soft-agar assay was performed as described previously (13).After 2 weeks or 1 month in culture, colonies were counted andimages were acquired at 5� magnification. Invasiveness wasexamined using a membrane invasion culture system (CorningLife Sciences) as described previously (13). Scratchwound assay isdetailed in Supplementary Materials and Methods.

Bersani et al.

Cancer Res; 76(20) October 15, 2016 Cancer Research6096




Western blot, FACS, and IHCWestern blot assay, MHC immunostaining for FACS analysis,

and IHC were performed as described previously (13, 15). Allantibodies used are listed in Supplementary Materials andMethods.

Chromatin immunoprecipitation assayChromatin immunoprecipitation (ChIP) assay was performed

as described previously (18). Antibodies and primers are listed inSupplementary Materials and Methods.

InhibitorsSelumetinib (AZD6244, Selleckchem) and lapatinib (GW-

572016, Selleckchem) were used at 1 mmol/L. DMSO was usedas a control.

In vivo tumorigenesis assayFor in vivo tumor growth, cells were resuspended in sterile PBS

and injected subcutaneously into the flank or in the tail vein offemale nu/nu mice (Charles River Laboratories). Tumor size wasmeasured as described previously (13). Conditional miR-22expression was induced in mice by administration of 1 mg/mLof doxycycline in the drinking water. All animal procedures wereapproved by the Ethical Committee of the University of Turin(Turin, Italy) and by the Italian Ministry of Health.

Statistical analysisTwo-tailed paired or unpaired Student t test was used to

evaluate statistical significance: NS, P > 0.05; �, P < 0.05; ��, P <0.01; ��,P< 0.001. Allmean values are expressed as SDor SEM, asspecified in the figure legends, and are derived from at least threeindependent experiments.

ResultsSmall RNA cloning and deep sequencing reveal a distinctmiRNA signature in rhabdomyosarcoma compared withnormal muscle

To identify novel miRNAs involved in rhabdomyosarcomapathogenesis, we profiled miRNA expression in 21 primaryhuman rhabdomyosarcoma and 4 normal muscles by NGS ofsmall RNA libraries. Unsupervised hierarchical clustering showeda strong distinction between muscles and tumors (Fig. 1A andSupplementary Fig. S1A; and Supplementary Table S1), revealinga clear rhabdomyosarcoma-related miRNA signature and indicat-ing a profound miRNA dysregulation in malignant tissues. How-ever, miRNA profiling did not distinguish among ERMS, ARMS,and PRMS, indicating that miRNAs cannot be predictive ofspecific subtypes. Notwithstanding, we identified few familiesoverrepresented in tumor samples (includingmiR-130a andmiR-335-5p; Fig. 1A), and not expressed in muscle. As expected,myomiRs (miR-1/206/133) were particularly underexpressed inrhabdomyosarcoma (Fig. 1A and Supplementary Fig. S1A; Sup-plementary Tables S1 and S2), confirming the good reliability ofour signature.

Interestingly, in the top 5muscle-enrichedmiRNAswe detectedalso miR-378 and miR-22 (Fig. 1A; Supplementary Table S2).While miR-378 has been recently associated to rhabdomyosar-coma pathogenesis (19), the role of miR-22 has remained so farunexplored.

miR-22 and -378 are muscle-enriched miRNAsWe first validated miR-22 and miR-378 expression levels using

alternative approaches. Northern blot analysis confirmed theenrichment of miR-22 and miR-378 in normal muscle, whileboth were barely detectable in primary rhabdomyosarcoma (Fig.1B). Furthermore, we assessed their presence and induction indifferent myogenic systems: differentiated satellite cells andC2C12myoblasts, miR-206–expressing rhabdomyosarcoma cells(13) and MyoD-reprogrammed 10T1/2 fibroblasts (15). In allmodels used and as previously reported for miR-22 in C2C12myoblasts (20), miR-22 and -378 were strongly induced duringdifferentiation (Fig. 1C–F), suggesting the involvement of bothmiRNAs in the myogenic program. Notably, for both miRNAsthere is complete homology in the mature sequence of humanand mouse, suggesting conserved functions. Interestingly, we didnot observe any substantial difference of expression between thetwo miRNAs in differentiated satellite cells and myoblasts. How-ever, miR-378 was the most upregulated miRNA in MyoD-repro-grammed 10T1/2 fibroblasts, whereas miR-22 was particularlyinduced in miR-206–expressing rhabdomyosarcoma cells. There-by, bothmiRNAs are involved in themyogenic program butmiR-22 may exert a critical role in rhabdomyosarcoma.

miR-22 acts as a potent oncosuppressor by interfering with thetransformed properties of rhabdomyosarcoma cells

To investigate the role of the two newly identified muscle-enriched miRNAs in rhabdomyosarcoma pathogenesis, we gen-erated rhabdomyosarcoma cells that conditionally expressedmiR-22 or -378 in a doxycycline-dependent manner. The systemwas tightly doxycycline-regulated with no expression in normalmedium, while doxycycline administration resulted in a robustmiRNA induction comparable with that observed in mouseskeletal muscle, especially for miR-22 (Supplementary Fig.S2A and S2B). miR-22 induction arrested proliferation of bothERMS (RD18) and ARMS (RH30) cells (Fig. 2A and B). Accord-ingly, cell-cycle distribution analysis showed a reduction in theS-phase and a concomitant accumulation in the G0–G1 and inG2–M phases of the cell cycle (Fig. 2C and D). Furthermore,miR-22 promoted apoptosis both in ERMS and ARMS cells (Fig.2E and F). These two aspects were confirmed by changes inexpression of proteins involved in cell proliferation and apo-ptosis (Fig. 2G). We next investigated the ability of miR-22 tointerfere with the transformed properties of rhabdomyosarcomacells. Long-term miR-22 induction significantly impairedanchorage-independent growth in soft-agar (Fig. 2H and I).Moreover, miR-22 expression impaired invasiveness in Matrigelassays (Fig. 2J and K).

Interestingly, miR-22 is hosted in the second exon of a longnoncoding RNA (lncRNA, MIR22HG). Thus, to rule out thepossibility that part of the oncosuppressor potential observedwas associated to the lncRNA,we generated a vector expressing thesame miR-22 precursor, but with the miR-22 seed sequencecompletelymutagenized. Using this construct, we did not observemature miR-22 production nor an effect on rhabdomyosarcomacell proliferation (Supplementary Fig. S2C and S2D). Finally, wealso explored the oncosuppressor potential of miR-378. Differ-ently from miR-22, miR-378 was, in our setting, less effective ininterfering with the transformed features of rhabdomyosarcomacells (Supplementary Fig. S2E–S2G and data not shown). Thus,we decided to further investigate the biological relevance of miR-22 only.

miR-22 Oncosuppressor Network in Rhabdomyosarcoma

www.aacrjournals.org Cancer Res; 76(20) October 15, 2016 6097




MyoD binds to the promoter region of miR-22 at the onset ofmyogenic differentiation and activates its transcription

The observation thatmiR-22was upregulated duringmyogeniccell differentiation suggested thatmuscle-regulatory factorsmightbe responsible for its activation. MyoD is themaster transcriptionfactor that governs the myogenic program (21). We thereforeexamined the core promoter of miR-22, consisting of 1,200 bpand we identified four predicted MyoD recognition sites, con-served in the mouse and human genes (Fig. 3A, black boxes).Thus, we set up a ChIP assay on MyoD-reprogrammed 10T1/2fibroblasts. As expected, MyoD and acetyl-histone H3 binding atthemiR-22 core promoter regions was observed only uponMyoDinduction (Fig. 3B and C). To verify whether MyoD bindingresulted in transcriptional activity, we generated three sensorconstructs (sensor 22.0, 22.1, and 22.2) containing 419, 586,and 1,089 bp upstream of pre-miR-22 fused to a luciferase

reporter. Upon MyoD induction, only sensors 22.1 and 22.2,containing two and four putative MyoD-binding sites, respective-ly, displayed strong transcriptional activity (Fig. 3D). Notwith-standing, no significant difference in luciferase induction wasobservedbetween the22.1 and22.2 constructs, indicating that thefirst two putative sites, located on the 22.1 fragment, were mainlyresponsible for MyoD binding. Accordingly, mutation of both E-boxes (IþII) of the 22.1 construct completely abolished luciferaseinduction (Fig. 3D), indicating that these two regulatory elementsare essential for MyoD transcriptional activity.

We previously showed that miR-206, another MyoD transcrip-tionally regulated myomiR (22), is able to force rhabdomyosarco-ma cells to resume differentiation (13). However, when weexplored the myogenic potential of miR-22, we did not observeany sign of terminal rhabdomyosarcoma differentiation (Supple-mentary Fig. S3A and S3B). Considering that MyoD promotes

Figure 1.

miR-22 and miR-378 aredownregulated in rhabdomyosarcoma(RMS) but are induced duringmyogenic differentiation. A, section ofthe expression dendrogram includingthe top 8 differentially expressedmiRNAs in human skeletal musclescompared with rhabdomyosarcomacells and tissues. Yellow, increasedexpression; black, reduced expression.B, Northern blot analysis on total RNAobtained from one muscle and onerhabdomyosarcoma sample. C and D,real-time PCR analysis in murinesatellite cells (C) and C2C12 cells (D)grown in proliferation (P) ordifferentiation medium (D). E and F,real-time PCR analysis in induciblemiR-206–expressing RD18 (NpBI-206)cells (E) and MyoD-expressing NIH10T1/2 (NpBI-MyoD) cells (F; IND,induced; NI, noninduced). Errorbars, SD.

Bersani et al.





myogenic differentiation by first arresting myoblast proliferationand subsequently bydriving terminal differentiation, it is likely thatmiR-22 is involved only in the MyoD-dependent cell-cycle arrest.

miR-22 abrogates tumor growth and disseminationThe ability ofmiR-22 to act as a potent oncosuppressor encour-

agedus to explore its therapeutic potential in vivo.Wefirst assessedwhether miR-22 was sufficient to prevent tumor growth. To thisaim, engineered rhabdomyosarcoma cells were injected intoimmunocompromised mice that were immediately treated withdoxycycline. While in control mice tumors grew rapidly, miR-22induction completely abolished rhabdomyosarcoma formation(Fig. 4A). Then, we evaluated the therapeutic potential of miR-22by inducing the miRNA only after the tumor became palpable.Also in this case, miR-22 expression resulted in potent inhibitionof rhabdomyosarcoma growth (Fig. 4B). The antitumorigenic

effect of miR-22 was confirmed by reduction of Ki67 stainingand by the presence of cleaved caspase-3–positive cells in treatedmice (Fig. 4C). Finally, we tested miR-22 ability to interfere withrhabdomyosarcoma cell dissemination. Tail vein injection ofnoninduced cells resulted in their rapid dissemination in lungsand kidneys (Fig. 4D and E, NI). Conversely, miR-22 inductionfully inhibited macro- and micrometastasis formation in bothorgans (Fig. 4D and E, IND). Immunohistochemical analysisconfirmed the presence of actively proliferating tumor cells onlyin tissues obtained from control mice (Fig. 4D and E).

Gene expression profiling of miR-22–expressingrhabdomyosarcoma cells identifies TACC1 and RAB5B as twocritical targets

To investigate the mechanism by which miR-22 interferedwith the transformed properties of rhabdomyosarcoma cells, we

Figure 2.

miR-22 interferes with thetransforming abilities ofrhabdomyosarcoma (RMS) cells. A andB, proliferation analysis of induciblemiR-22–expressing RD18 (A) and RH30(B) NpBI-22 cells (IND, induced; NI,noninduced). C and D, cell-cycledistribution of RD18 (C) and RH30 (D)NpBI-22 cells (induced andnoninduced). E and F, apoptosisassessment of RD18 (E) and RH30 (F)NpBI-22 cells (induced andnoninduced). G, Western blotanalysis in RD18 and RH30 NpBI-22cells (induced and noninduced). Hand I,quantification and representativeimages of soft-agar growth of RD18 (H)and RH30 (I) NpBI-22 cells (inducedand noninduced). J andK, invasivenessassessment of RD18 (J) and RH30 (K)NpBI-22 cells (induced andnoninduced). Error bars, SEM.� , P < 0.05; �� , P < 0.01; �� , P < 0.001(t test).






compared gene expression profiling of RD18 NpBI-22 induced(IND) cells with noninduced (NI) cells. Unsupervised hierarchi-cal clustering resulted in a dendrogram with two clearly distinctbranches separating miR-22–expressing cells from controls (Sup-plementary Fig. S4A). In particular, ANOVA analysis revealed atotal of 494 significantly modulated transcripts, among which263 were downregulated in the induced condition. To identifypossible miR-22 target genes, we analyzed the downmodulatedtranscripts using the EIMMo miRNA target prediction server(Supplementary Table S3). Among them, Transforming AcidicCoiled-Coil Containing Protein 1 (TACC1) andRAB5B resulted ofparticular interest. TACC1 is indeed a cancer-related gene locatedon chromosome 8p11, amplified in breast cancer (23) andtranslocated in glioblastoma (24). RAB5, instead, is a smallGTPase involved in intracellular trafficking and migration (25).Notably, TACC1 30UTR contains twomiR-22 MREs, while RAB5B

30UTR includes three of them (Supplementary Fig. S4D), suggest-ing that both genes could be directly and efficiently downmodu-lated bymiR-22. Western blot analysis confirmed a strong TACC1and RAB5B downregulation uponmiR-22 induction, both in vitroand in vivo (Supplementary Fig. S4B and S4C). Moreover, sensorvectors expressing themutagenized or thewild-typemiR-22MREsofTACC1 andRAB5B30UTRs showed that both geneswere indeedmiR-22 direct targets (Supplementary Fig. S4E and S4F).

TACC1 and RAB5B control the transformed properties ofrhabdomyosarcoma cells

To study the functional relevance of TACC1 in rhabdomyosar-coma pathogenesis, we first tested the effects of its downregula-tion in rhabdomyosarcoma cells. Interestingly, TACC1 silencing(Fig. 5A) impaired cell proliferation (Fig. 5B, C, and F), promotedapoptosis (Fig. 5D and F), and inhibited soft-agar growth of both

Figure 3.

miR-22 is transcriptionally regulated byMyoD. A, schematic representation ofmiR-22 promoter region. The putativebinding sites for MyoD are indicatedbyblack boxes.B andC,ChIP analysis ofmiR-22 promoter regions in NIHNpBI-MyoD cells (IND, induced; NI,noninduced). Two independent oligopairs were used for miR-22 promoteramplification (oligos A and oligos B).MCK and IgH enhancers were used aspositive and negative controls,respectively. Recruitment was relativeto normal rabbit IgG. D, dual luciferaseassay in NIH NpBI-MyoD cells (inducedand noninduced) transfected withthe indicated luciferase reporterconstructs. Error bars, SD. � , P < 0.05;�� , P < 0.01; �� , P < 0.001 (t test).

Bersani et al.





ERMS and ARMS cells (Fig. 5E). Conversely, TACC1 overexpres-sion (Fig. 5G) partially rescued the anchorage-independentgrowth ability of miR-22–expressing rhabdomyosarcoma cells(Fig. 5H).

RAB5 expression has been previously linked with tumor cellmigration and invasion (26, 27). However, RAB5 is encoded bythree distinct paralogs: RAB5A, RAB5B, and RAB5C. Interestingly,RAB5B is the prevalent isoform expressed in cells of both rhab-domyosarcoma subtypes (Fig. 6A and B). Thus, we silencedRAB5B in rhabdomyosarcoma cells (Fig. 6C) and exploredwheth-er RAB5B was involved in rhabdomyosarcoma cell motility.Wound-healing assay showed that RAB5B silencing delayedwound closure at the evaluated time point (Fig. 6D and E),whereas RAB5B overexpression (Fig. 6F) enhanced migrationtoward the wounded area of miR-22–expressing rhabdomyosar-coma cells (Fig. 6G and H). Altogether, our data support a model

in which the oncosuppressor role of miR-22 in rhabdomyosar-coma cells is, at least in part, due to TACC1 and RAB5Bdownregulation.

miR-22 prevents intrinsic resistance to MEK inhibition byintercepting ERBB3 upregulation

Comprehensive genomic analyses revealed that the RAS path-way is frequently mutated in ERMS tumors (8–10). Although thepharmacologic inhibition of constitutively active RAS proteinsremains challenging (28), small-molecule inhibitors directedagainst the downstream ERK signaling offer an alternative RASpathway–targeted strategy. Nevertheless, the efficacy of MEKinhibitors in cancer patients is modest (29). This partial responsehas been frequently associated to the activation of compensatoryfeedback pathways involved in primary resistance (30, 31). Nota-bly, RD18 cells harbor the constitutive active formofNRAS (NRAS

Figure 4.

miR-22 blocks rhabdomyosarcoma(RMS) tumor growth and dissemination.A and B, tumor growth curve in nudemice subcutaneously injected with RD18NpBI-22 cells. A, half of the mice (n ¼ 6)were given doxycycline starting at thetime of injection (IND, induced), whilethe rest remained untreated (NI,noninduced). B, doxycycline was givento half of the mice (n ¼ 6) only whentumors become palpable (black arrow).C, immunohistochemical analysis oftumors recovered at the end of thetreatment explained in B. D and E,representative images of whole mountlungs (D, top) and kidneys (E, top)recovered 3 months after tail veininjection of RD18 NpBI-22 cells in nudemice (induced and noninduced). Lowerpanels show immunohistochemicalanalysis on sections from the tissuesdescribed above. Error bars, SEM.






Q61H; ref. 32) and are relatively resistant to MEK inhibition(Fig. 7A; ref. 30). Considering the strong oncosuppressor poten-tial of miR-22, we explored whether this miRNA could also play arole in preventing resistance to the MEK inhibitor selumetinib.Intriguingly, miR-22 induction in combination with MEK inhi-bition potently blocked RD18 cell growth, enhancing the effectsof the single treatment alone (Fig. 7A and C). Interestingly, arecent synthetic lethal screening approach revealed that a com-mon mechanism of primary resistance to MEK inhibitors impli-cates ERBB3 upregulation (33). Also in RD18 cells, selumetinibtreatment resulted in a strong ERBB3 induction (Fig. 7B). Con-versely, in the presence of miR-22, selumetinib was unable toupregulate ERBB3 (Fig. 7B), suggesting that miR-22 was interfer-ing with ERBB3 expression. Accordingly, ERBB3 transcript con-tains a functional miR-22 MRE in the 30UTR (Fig. 7D) and thusmiR-22 can immediately intercept ERBB3 expression upon selu-metinib treatment. Finally, the combination treatment based on

selumetinib plus lapatinib also enhanced selumetinib-mediatedsoft-agar growth inhibition (Fig. 7C), further confirming thefunctional relevance of ERBB3 pathway activation in drivingMEKinhibitor resistance.

DiscussionIn this work, we performed miRNA NGS on primary rhabdo-

myosarcoma samples, rhabdomyosarcoma cell lines, and normalskeletalmuscles. ThemiRNA signature revealed a clear distinctionbetween muscle and tumor specimens but, in contrast to aprevious study (34), we were unable to distinguish the threehistologic categories included in our samples (ERMS, ARMS, andPRMS), suggesting a common alteration in miRNA expressionprofile independently on rhabdomyosarcoma subtype.

However, our miRNA expression profiling revealed that miR-378 and miR-22, together with other previously characterized

Figure 5.

TACC1 promotes rhabdomyosarcoma(RMS) cell proliferation, anchorage-independent growth, and protectsrhabdomyosarcoma cells fromapoptosis. A, Western blot analysis inRD18 and RH30 cells expressing theshRNA control (shCTRL) or TACC1-directed shRNA (shTACC1). B and C,proliferation analysis of RD18 (B) andRH30 (C) cells described in A. D,apoptosis assessment of RD18 and RH30cells described inA.E,quantification andrepresentative images of soft-agargrowth of RD18 and RH30 cellsdescribed inA after 2weeks in culture. F,Western blot analysis in RD18 and RH30cells described in A. G, Western blotanalysis in RD18 NpBI-22 cellsexpressing GFP (CTRL) or TACC1. H,quantification and representativeimages of soft-agar growth of RD18NpBI-22 cells described in G andtreated with doxycycline (IND, induced)for 2 weeks. Error bars, SEM. � , P < 0.05;�� , P < 0.01; �� , P < 0.001 (t test).

Bersani et al.





myomiRs, were markedly underrepresented in rhabdomyosar-coma compared with normal muscle tissue. While miR-378downregulation in rhabdomyosarcoma has been recentlyreported (19), the role of miR-22 has remained so far unex-plored. The biological significance of miR-22 in cancer is stillcontroversial and seems to be context specific. While it isgenerally considered as an oncosuppressor (35–38), two recentarticles support its oncogenic potential (39, 40). Although miR-22 is ubiquitously expressed, it is particularly enriched incardiac muscle, where it is involved in cardiac hypertrophyand remodeling (41, 42). According to our data, miR-22 is alsoabundant in skeletal muscle and the muscle-specific transcrip-tion factor MyoD plays a critical role in regulating its expressionin myogenic cells. While genome-wide analysis of MyoD-bind-ing regions in C2C12 myoblasts indicated its presence in thepromoter of four miRNAs, including miR-22 (21), here we

identified the two paired sites upstream of miR-22, critical forboth MyoD binding and transcription. As reported by others(43), MyoD is not functional in rhabdomyosarcoma despite itsability to associate with coactivators and to bind to DNA, andthis could explain why miR-22 expression is impaired in thistumor type. MyoD acts at two levels in myogenesis regulatingfirst the block of cell proliferation and then muscle-specificgene expression (44). The mediators of the former effect are stillpoorly understood. Our data suggest that the antiproliferativeaction of MyoD is accomplished, at least in part, post-tran-scriptionally by miR-22. Indeed, when ectopically expressed inrhabdomyosarcoma cells of both embryonal and alveolar his-tology, miR-22 showed a prominent oncosuppressor potentialby impairing cell proliferation, invasiveness, and significantlyincreasing apoptosis. Besides its strong in vitro effect, miR-22also displayed a remarkable antitumor effect in vivo, leading to a

Figure 6.

RAB5B controls rhabdomyosarcoma(RMS) cell migration.A andB, real-timePCR analysis of RAB5 paralogtranscripts in RD18 (A) and RH30 (B)cells. C, Western blot analysis in RD18cells transfected with siRNA control(siCTRL) or with RAB5-directed siRNAs(siRAB5A, B, or C). Representativeimages (D) and quantification (E) ofwound closure of RD18 cells transfectedwith siCTRL or siRAB5B. F, Westernblot analysis in RD18 NpBI-22 cellsexpressing GFP (CTRL) or RAB5B.Representative images (G) andquantification (H) of wound closure ofRD18 NpBI-22 cells maintained inpresence of doxycycline (IND, induced)and expressing GFP (CTRL) or RAB5B.Error bars, SEM. � , P < 0.05 (t test).






significant reduction of rhabdomyosarcoma growth and met-astatic potential.

By gene expression analysis in miR-22–expressing RD18cells, we identified TACC1 and RAB5B as promising miR-22–regulated candidates. TACC1 and RAB5, both acting oncytoskeleton dynamics, have been previously shown to beinvolved in cell transformation and in tumor dissemination.TACC1 was first identified as the sole coding sequence withinthe 8p11 breast cancer–associated amplicon and capable oftransforming mouse fibroblasts by itself (23). Interestingly, asignificant percentage of ERMS (up to 92%) displays gain ofthe entire chromosome 8 (8–11), suggesting a possible onco-genic role for genes harbored in this chromosome. Conversely,RAB5 family members are critical trafficking molecules thatdirectly influence several aspects of cell migration (26, 27).Accordingly, modulation of either target genes influenced thetransformed properties of rhabdomyosarcoma cells. WhileTACC1 was mainly involved in the regulation of rhabdomyo-sarcoma cell growth, RAB5B prevalently controlled rhabdo-myosarcoma cell motility. Interestingly, RAB5A is also a pre-dicted target of miR-206, and considering that all RAB5 mem-bers play a redundant role in actin cytoskeleton dynamics, wecannot rule out the possibility that more complex circuitsinvolving myomiRs and RAB5 genes are dysregulated in rhab-domyosarcoma pathogenesis.

As a single miRNA can potentially regulate hundreds ofdifferent transcripts, other targets are likely to contribute tomiR-22 therapeutic efficacy. This concept is particularly rele-vant considering the widely accepted view that cancer is aDarwinian disease (31). Regrettably, the targeted therapyapproach, less toxic than conventional chemotherapy, oftengives rise to adaptive responses that result in resistance (31). Inthis context, a combination of a target therapy with a pleotropicmiRNA with therapeutic potential could be beneficial in pre-venting mechanisms of resistance. We thus explored the func-tional relevance of miR-22 action in this setting. Interestingly,RD18 cells that harbor a constitutive active form of NRAS aremodestly responsive to selumetinib treatment and here we haveshown that this is in part related to a feedback mechanism ofcompensation based on ERBB3 upregulation. Notably, ERBB3was not present in our list of downregulated genes upon miR-22 induction, although it includes one functional MRE. Thisapparent discrepancy is based on the fact the profiling wasperformed in cells growing in medium without drugs. Indeed,ERBB3 upregulation was observed only upon selumetinibtreatment, and in this condition, the concomitant presence ofmiR-22 efficiently interfered with the mechanism of resistance.Accordingly, we have shown that the combination of drugs(selumetinib þ lapatinib) was more effective than single treat-ments. However this pharmacologic strategy could still be

Figure 7.

miR-22 intercepting ERBB3counteracts the resistance to MEKinhibitors. A, proliferation analysis ofRD18 NpBI-22 cells induced (IND) ornoninduced (NI) and treated withselumetinib (SEL) or combination.B,Western blot analysis in RD18 NpBI-22 cells induced or noninduced andtreated with lapatinib (LAP),selumetinib (SEL), or combinations forthree days. C, quantification andrepresentative images of soft-agargrowth of RD18 NpBI-22 cellsdescribed in B. D, dual luciferase assayin 293T cells transfected with theindicated luciferase reporterconstructs (WT, wild-type; MUT,mutated) along with miR-22. Errorbars, SEM. � , P < 0.05; �� , P < 0.01;�� , P < 0.001 (t test).

Bersani et al.





inadequate to prevent the emergence of further resistance,especially in highly heterogeneous ERMS.

For this reasons, future therapeutic strategies against rhabdo-myosarcoma could take advantage of using pleotropic oncosup-pressor miRNAs (13, 19, 45–50), among which we now proposemiR-22.

Overall, our miRNA NGS and profiling reveal a novel miRNAnetwork in whichmiR-22 plays a critical role by acting atmultiplelevels of regulation. Although drugs directed against individualmiR-22 targets may provide useful alternative strategies in rhab-domyosarcoma, our data suggest that, once the problem ofmiRNA delivery will be overcome, miR-22 could represent aneffective therapeutic option aloneor in combinationwith targetedagents in primary tumors, distant metastases and to prevent drug-induced primary resistance.

Disclosure of Potential Conflicts of InterestT. Tuschl is a consultant/advisory board member for Alnylam Pharmaceu-

ticals and Regulus Therapeutics. No potential conflicts of interest were disclosedby the other authors.

Authors' ContributionsConception and design: F. Bersani, M.F. Lingua, R. TaulliDevelopment of methodology: F. Bersani, T. Tuschl, R. TaulliAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): M.F. Lingua, D. Morena, V. Foglizzo, S. Miretti,R. Chiarle, S. Singer, M. Ladanyi

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): F. Bersani, M.F. Lingua, D. Morena, V. Foglizzo,S. Miretti, G. Carr�a, A. Morotti, U. Ala, P. Provero, T. TuschlWriting, review, and/or revision of the manuscript: F. Bersani, M.F. Lingua,D. Morena, A. Morotti, T. Tuschl, C. Ponzetto, R. TaulliAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): F. Bersani, L. LanzettiStudy supervision: F. Bersani, R. Taulli

AcknowledgmentsThe authors wish to thank PavelMorozov (Rockefeller University, New York,

NY) for bioinformatic support and Maria Stella Scalzo (University of Turin,Turin, Italy) for help with immunohistochemical analysis. The support of theFondazione Ricerca Molinette Onlus is gratefully acknowledged.

Grant SupportThis work was supported by the Italian Association for Cancer Research

(AIRC Project number IG12089 to C. Ponzetto; AIRC Start-up Project number15405 to R. Taulli; AIRC Project number IG15180 to L. Lanzetti), EMBO ShortTerm Fellowship (ASTF No. 336-07 to F. Bersani) and Regione Piemonte(IMMONC Project; to C. Ponzetto).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received March 11, 2016; revised July 4, 2016; accepted July 28, 2016;published OnlineFirst August 28, 2016.

References1. Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell

2009;136:215–33.2. Fabian MR, Sonenberg N, Filipowicz W. Regulation of mRNA translation

and stability by microRNAs. Annu Rev Biochem 2010;79:351–79.3. Krol J, Loedige I, Filipowicz W. The widespread regulation of microRNA

biogenesis, function and decay. Nat Rev Genet 2010;11:597–610.4. Cacchiarelli D, Martone J, Girardi E, CesanaM, Incitti T, MorlandoM, et al.

MicroRNAs involved in molecular circuitries relevant for the duchennemuscular dystrophy pathogenesis are controlled by the dystrophin/nNOSpathway. Cell Metab 2010;12:341–51.

5. Nana-Sinkam SP, Croce CM. MicroRNA regulation of tumorigenesis,cancer progression and interpatient heterogeneity: towards clinical use.Genome Biol 2014;15:445.

6. Hayes J, Peruzzi PP, Lawler S. MicroRNAs in cancer: biomarkers, functionsand therapy. Trends Mol Med 2014;20:460–9.

7. Shern JF, Yohe ME, Khan J. Pediatric Rhabdomyosarcoma. Crit Rev Oncog2015;20:227–43.

8. Chen X, Stewart E, Shelat AA, Qu C, Bahrami A, Hatley M, et al. Targetingoxidative stress in embryonal rhabdomyosarcoma. Cancer Cell 2013;24:710–24.

9. Shern JF, Chen L, Chmielecki J, Wei JS, Patidar R, Rosenberg M, et al.Comprehensive genomic analysis of rhabdomyosarcoma reveals a land-scape of alterations affecting a common genetic axis in fusion-positive andfusion-negative tumors. Cancer Discov 2014;4:216–31.

10. Seki M, Nishimura R, Yoshida K, Shimamura T, Shiraishi Y, Sato Y, et al.Integrated genetic and epigenetic analysis defines novel molecular sub-groups in rhabdomyosarcoma. Nat Commun 2015;6:7557.

11. Williamson D, Missiaglia E, de Reyni�es A, Pierron G, Thuille B, PalenzuelaG, et al. Fusion gene-negative alveolar rhabdomyosarcoma is clinically andmolecularly indistinguishable fromembryonal rhabdomyosarcoma. J ClinOncol 2010;28:2151–8.

12. Keller C, Guttridge DC. Mechanisms of impaired differentiation in rhab-domyosarcoma. FEBS J 2013;280:4323–34.

13. Taulli R, Bersani F, Foglizzo V, Linari A, Vigna E, Ladanyi M, et al. Themuscle-specific microRNA miR-206 blocks human rhabdomyosarcomagrowth in xenotransplantedmice by promotingmyogenic differentiation. JClin Invest 2009;119:2366–78.

14. Taulli R, Foglizzo V, Morena D, Coda DM, Ala U, Bersani F, et al. Failure todownregulate the BAF53a subunit of the SWI/SNF chromatin remodelingcomplex contributes to the differentiation block in rhabdomyosarcoma.Oncogene 2014;33:2354–62.

15. CodaDM, LinguaMF,MorenaD, Foglizzo V, Bersani F, AlaU, et al. SMYD1and G6PD modulation are critical events for miR-206-mediated differen-tiation of rhabdomyosarcoma. Cell Cycle 2015;14:1389–402.

16. La�eM, AhnEH,MercadoGE,Chuai S, EdgarM, Pawel BR, et al. Global geneexpression profiling of PAX-FKHR fusion-positive alveolar and PAX-FKHRfusion-negative embryonal rhabdomyosarcomas. J Pathol 2007;212:143–51.

17. Hafner M, Landgraf P, Ludwig J, Rice A, Ojo T, Lin C, et al. Identification ofmicroRNAs and other small regulatory RNAs using cDNA library sequenc-ing. Methods 2008;44:3–12.

18. Palacios D, Mozzetta C, Consalvi S, Caretti G, Saccone V, Proserpio V, et al.TNF/p38a/polycomb signaling to Pax7 locus in satellite cells links inflam-mation to the epigenetic control of muscle regeneration. Cell Stem Cell2010;7:455–69.

19. Megiorni F, Cialfi S, McDowell HP, Felsani A, Camero S, Guffanti A, et al.Deep Sequencing the microRNA profile in rhabdomyosarcoma revealsdown-regulation of miR-378 family members. BMC Cancer 2014;14:880.

20. Marzi MJ, Puggioni EMR, Dall'Olio V, Bucci G, Bernard L, Bianchi F, et al.Differentiation-associated microRNAs antagonize the Rb-E2F pathway torestrict proliferation. J Cell Biol 2012;199:77–95.

21. Cao Y, Kumar RM, Penn BH, Berkes CA, Kooperberg C, Boyer LA, et al.Global and gene-specific analyses showdistinct roles forMyodandMyog ata common set of promoters. EMBO J 2006;25:502–11.

22. Rao PK, Kumar RM, Farkhondeh M, Baskerville S, Lodish HF. Myogenicfactors that regulate expression of muscle-specific microRNAs. Proc NatlAcad Sci U S A 2006;103:8721–6.

23. Still IH, Hamilton M, Vince P, Wolfman A, Cowell JK. Cloning ofTACC1, an embryonically expressed, potentially transforming coiledcoil containing gene, from the 8p11 breast cancer amplicon. Oncogene1999;18:4032–8.

24. Singh D, Chan JM, Zoppoli P, Niola F, Sullivan R, Castano A, et al.Transforming fusions of FGFR and TACC genes in human glioblastoma.Science 2012;337:1231–5.






25. Lanzetti L, Palamidessi A, Areces L, Scita G, Di Fiore PP. Rab5 is a signallingGTPase involved in actin remodelling by receptor tyrosine kinases. Nature2004;429:309–14.

26. Palamidessi A, Frittoli E, Garr�e M, Faretta M, Mione M, Testa I, et al.Endocytic trafficking ofRac is required for the spatial restrictionof signalingin cell migration. Cell 2008;134:135–47.

27. Liu S, Chen X, Zheng H, Shi S, Li Y. Knockdown of Rab5a expressiondecreases cancer cell motility and invasion through integrin-mediatedsignaling pathway. J Biomed Sci 2011;18:58.

28. Patricelli MP, Janes MR, Li L-S, Hansen R, Peters U, Kessler LV, et al.Selective inhibition of oncogenicKRAS output with small molecules tar-geting the inactive state. Cancer Discov 2016;6:316–29.

29. Samatar AA, Poulikakos PI. Targeting RAS-ERK signalling in cancer: pro-mises and challenges. Nat Rev Drug Discov 2014;13:928–42.

30. Renshaw J, Taylor KR, Bishop R, ValentiM,DeHaven Brandon A,Gowan S,et al. Dual blockade of the PI3K/AKT/mTOR (AZD8055) and RAS/MEK/ERK (AZD6244) pathways synergistically inhibits rhabdomyosarcoma cellgrowth in vitro and in vivo. Clin Cancer Res 2013;19:5940–51.

31. Misale S, Arena S, Lamba S, SiravegnaG, Lallo A,Hobor S, et al. Blockade ofEGFR and MEK intercepts heterogeneous mechanisms of acquired resis-tance to anti-EGFR therapies in colorectal cancer. Sci Transl Med 2014;6:224ra26.

32. Chardin P, Yeramian P, Madaule P, Tavitian A. N-ras gene activation in theRD human rhabdomyosarcoma cell line. Int J Cancer 1985;35:647–52.

33. Sun C, Hobor S, Bertotti A, Zecchin D, Huang S, Galimi F, et al. Intrinsicresistance to MEK inhibition in KRAS mutant lung and colon cancerthrough transcriptional induction of ERBB3. Cell Rep 2014;7:86–93.

34. Subramanian S, Lui WO, Lee CH, Espinosa I, Nielsen TO, Heinrich MC,et al. MicroRNA expression signature of human sarcomas. Oncogene2008;27:2015–26.

35. Zhang J, Yang Y, Yang T, Liu Y, Li A, Fu S, et al. microRNA-22, down-regulated in hepatocellular carcinoma and correlated with prognosis,suppresses cell proliferation and tumourigenicity. Br J Cancer 2010;103:1215–20.

36. Xu D, Takeshita F, Hino Y, Fukunaga S, Kudo Y, Tamaki A, et al. miR-22represses cancer progression by inducing cellular senescence. J Cell Biol2011;193:409–24.

37. TsuchiyaN, IzumiyaM,Ogata-KawataH,OkamotoK, Fujiwara Y,NakaiM,et al. Tumor suppressor miR-22 determines p53-dependent cellularfate through post-transcriptional regulation of p21. Cancer Res 2011;71:4628–39.

38. Jiang X, Hu C, Arnovitz S, Bugno J, Yu M, Zuo Z, et al. miR-22 has a potentanti-tumour role with therapeutic potential in acute myeloid leukaemia.Nat Commun 2016;7:11452.

39. SongSJ, ItoK,AlaU,Kats L,WebsterK, SunSM,et al. TheoncogenicmicroRNAmiR-22 targets theTET2 tumor suppressor topromotehematopoietic stemcellself-renewal and transformation. Cell Stem Cell 2013;13:87–101.

40. Song SJ, Poliseno L, Song MS, Ala U, Webster K, Ng C, et al. MicroRNA-antagonism regulates breast cancer stemness and metastasis via TET-family-dependent chromatin remodeling. Cell 2013;154:311–24.

41. Huang Z-P, Chen J, Seok HY, Zhang Z, Kataoka M, Hu X, et al. MicroRNA-22 regulates cardiac hypertrophy and remodeling in response to stress. CircRes 2013;112:1234–43.

42. Zhao H, Wen G, Wen G, Huang Y, Yu X, Chen Q, et al. MicroRNA-22regulates smooth muscle cell differentiation from stem cells by targetingmethyl CpG-binding protein 2. Arterioscler Thromb Vasc Biol 2015;35:918–29.

43. Tapscott SJ, ThayerMJ,WeintraubH.Deficiency in rhabdomyosarcomas ofa factor required for MyoD activity and myogenesis. Science 1993;259:1450–3.

44. de la Serna IL, Roy K, Carlson KA, Imbalzano AN. MyoD can induce cellcycle arrest but not muscle differentiation in the presence of dominantnegative SWI/SNF chromatin remodeling enzymes. J Biol Chem2001;276:41486–91.

45. WangH,GarzonR, SunH, LadnerKJ, SinghR,Dahlman J, et al.NF-kappaB-YY1-miR-29 regulatory circuitry in skeletal myogenesis and rhabdomyo-sarcoma. Cancer Cell 2008;14:369–81.

46. Yan D, Dong XDE, Chen X, Wang L, Lu C, Wang J, et al. MicroRNA-1/206targets c-Met and inhibits rhabdomyosarcoma development. J Biol Chem2009;284:29596–604.

47. Rao PK,Missiaglia E, Shields L, Hyde G, Yuan B, Shepherd CJ, et al. Distinctroles for miR-1 and miR-133a in the proliferation and differentiation ofrhabdomyosarcoma cells. FASEB J 2010;24:3427–37.

48. Balkhi MY, Balkhi MY, Iwenofu OH, Bakkar N, Ladner KJ, Chandler DS,et al. miR-29 acts as a decoy in sarcomas to protect the tumor suppressorA20 mRNA from degradation by HuR. Sci Signal 2013;6:ra63.

49. Diao Y, Guo X, Jiang L, Wang G, Zhang C, Wan J, et al. miR-203, a tumorsuppressor frequently down-regulated by promoter hypermethylation inrhabdomyosarcoma. J Biol Chem 2014;289:529–39.

50. Vella S, Pomella S, Leoncini PP, Colletti M, Conti B, Marquez VE, et al.MicroRNA-101 is repressed by EZH2 and its restoration inhibits tumorigenicfeatures in embryonal rhabdomyosarcoma. Clin Epigenetics 2015;7:82.


Bersani et al.




2016;76:6095-6106. Published OnlineFirst August 28, 2016.Cancer Res Francesca Bersani, Marcello Francesco Lingua, Deborah Morena, et al. Therapeutic Potential in RhabdomyosarcomaDeep Sequencing Reveals a Novel miR-22 Regulatory Network with

Updated version

10.1158/0008-5472.CAN-16-0709doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2016/08/26/0008-5472.CAN-16-0709.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/76/20/6095.full#ref-list-1

This article cites 50 articles, 18 of which you can access for free at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/76/20/6095To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-16-0709

http://cancerres.aacrjournals.org/content/suppl/2016/08/26/0008-5472.CAN-16-0709.DC1

http://cancerres.aacrjournals.org/content/76/20/6095.full#ref-list-1

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/76/20/6095


deep sequencing reveals a novel mir-22 regulatory network with therapeutic potential ... ·...

Documents